# Misata: synthetic data library

> Misata: synthetic data library is an open-source MCP server for databases. Generate multi-table synthetic data that provably meets the rules you declare. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** CLI, self-hosted
- **Website:** https://www.misata.studio/docs
- **Repository:** https://github.com/rasinmuhammed/misata
- **MCP Registry name:** io.github.rasinmuhammed/misata
- **Categories:** databases, developer tools, analytics and data
- **Listing updated:** 2026-11-15

## About

Misata starts from a statement of what must be true: tables and keys, CHECK rules, a monthly revenue curve, a refund rate, a 50/35/15 channel split, legal order of subscription states. It generates rows that satisfy all of it, refuses before generating when two declarations contradict each other, and can write an attestation that anyone can re-run in DuckDB.

Input can be SQL DDL, a YAML schema, a live Postgres or SQLite database, Django models, or a single sentence. Output goes to CSV, JSON Lines, Parquet, DuckDB, Postgres or SQLite. Within one Misata version, the same schema and seed give the same rows on every machine. A `mimic` mode fits real rows, and its output is not anonymous.

The library is MIT-licensed Python and needs no real data, trained model or API key. The MCP server comes from the PyPI package and runs over stdio. The site also sells ready-made datasets from $5 and a credit-based Studio, still in beta.

## Features

- **Declared facts hold:** Rows are generated to satisfy the stated rules and an explain step reports conflicts first.
- **DuckDB attestation:** Pins every file by SHA-256 and ships a verify script for each invariant.
- **Many inputs:** Starts from DDL, YAML, a database, Django models or a sentence.
- **Many outputs:** Writes files or seeds Postgres and SQLite directly.
- **Reproducible seeds:** Same schema, seed and version give identical rows.
- **Pipeline solving:** Can generate data so a dbt or SQL pipeline hits chosen targets.

## Good for

- Seed a dev or CI database from its DDL
- Test dbt models against data with known correct answers
- Build a demo dataset with believable skew
- Create evaluation sets with verified answers
- Publish a verified dataset to Hugging Face

## Install

### Claude Code (Terminal)

```sh
claude mcp add misata -- uvx misata
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "misata": {
      "command": "uvx",
      "args": [
        "misata"
      ]
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "misata": {
      "command": "uvx",
      "args": [
        "misata"
      ]
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "misata": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "misata"
      ]
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "misata": {
      "command": "uvx",
      "args": [
        "misata"
      ]
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "misata": {
      "command": "uvx",
      "args": [
        "misata"
      ]
    }
  }
}
```

## FAQ

### Does it need real data?

No. It needs no real data, trained model or API key.

### Is mimic output anonymous?

No. Mimic fits real rows, so it is not anonymous or privacy-safe.

### Can it fill Postgres directly?

Yes, with `misata seed` against a Postgres URL.

### How is it different from Faker?

Faker makes individual values. Misata declares cross-table rules and proves afterwards that they hold.

### Is Misata: synthetic data library free?

Yes. Misata: synthetic data library is open source, so you can use and self-host it for free.

### How do I add Misata: synthetic data library to Claude Code?

Run this command in your terminal: claude mcp add misata -- uvx misata

### What are the best alternatives to Misata: synthetic data library?

The closest alternatives to Misata: synthetic data library are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [GitHub](https://ai-powered.com/mcp/github-2): GitHub is an open-source MCP server for developer tools.
- [Serena MCP: the IDE for your agent](https://ai-powered.com/mcp/serena-mcp-the-ide-for-your-agent): Serena MCP: the IDE for your agent is an open-source MCP server for developer tools.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Agent-Native Analytics](https://ai-powered.com/mcp/agent-native-analytics): Agent-Native Analytics is an open-source MCP server for analytics and data.
- [Ouroboros](https://ai-powered.com/mcp/ouroboros): Ouroboros is an open-source MCP server for developer tools.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.

---
Source: https://ai-powered.com/mcp/misata-synthetic-data-library
